Re: Any performance benefits in going to db_16k_cache_size or db_32k_cache_size

I would think that the main benefit of the larger blocksize is to address table rows that wouldn't otherwise fit in one 8k block, so as to avoid row chaining or row migration. Perhaps having a larger block would also equate to less I/O though.

Don.

On 1/9/07, arun chakrapani rao <arunchakrapanirao_at_gmail.com> wrote:
> Hi all,
> I am currently working on a 1.5T DSS running on 8k blocksize.
> This db is 90% of the time doing sequential reads.
> Was wondering if we go for a 16k or 32k blocksize just on index alone
> would there be any benefits in performance.
> Has anybody impletemented this and seen any performance benefits,
> Please do share your experience.
> I was trying to do some bench marking here for one of the queries with
> 8,16 and 32k index tablespaces.
> All I am seeing is the logical reads going down by half but the
> elapsed time for these queries are still the same.
> Now what am i missing?
> Please do share your thoughts.
>
> --
> thanks in advance.
> Arun
> --
> http://www.freelists.org/webpage/oracle-l
>
>
>

--
http://www.freelists.org/webpage/oracle-l